Back to blog

Who signs off when your AI system gets it wrong?

Thắng Đoàn
Thắng Đoàn

Your AI model made a bad call that hurt a customer. Nobody can say who approved it, who monitors it, or who answers for it.

Governance is not the gap, ownership is

The process is fine; the ownership behind it is missing. Most enterprises govern AI the way they would any other IT rollout, cio.com reports: deploy dashboards, institute policies, conduct periodic reviews. The article argues that having the right tools matters, but IT executives should worry more about rethinking workflows and establishing ownership. The result of ignoring this is inconsistent use, fragmented controls, and limits on scale. In short, the process is fine; the ownership behind it is missing.

CIOs own the blame but not the budget

The authority gap is measurable. A Thoughtworks survey of 3,200 CIOs, reported by Morningstar, finds 91% of North American CIOs say AI adoption is outpacing governance. The survey points to a widening gap between authority and accountability as AI tools spread through the enterprise. Fragmented budget ownership leaves accountability and control disjointed. You can be answerable for a failure you had no authority to prevent.

Is a chief AI officer the fix?

Writing on Forbes, Mark Thirlwell of BSI notes the problem is not new. President Truman's "the buck stops here" joined power to responsibility, and AI is forcing the same pairing now. Every business function touches AI, from payroll to candidate review, so accountability is becoming a board-level concern. But Thirlwell is blunt: as things stand, no single accountability model exists. A new title alone does not create ownership; it just relocates the vacuum.

What this does not solve

None of these sources prove that a chief AI officer, or any specific structure, reduces AI failures. The evidence describes who feels responsible, not what actually works. Surveys of CIO sentiment cannot tell you whether a governance model prevents harm. Treat these findings as a map of the gap, not a validated way to close it.

If one of your AI systems failed today, could you name the person who signs off on it?
Does your CIO hold accountability for AI outcomes without matching budget authority?

List your three riskiest AI deployments and write the owner's name next to each. Any blanks are your answer.

Sources

CIO: AI governance's real gap is accountability

Morningstar: 91% of CIOs say AI outpaces governance

Forbes: Is a chief AI officer the answer?

Share:

Recommended for you

Enjoyed this article?

Subscribe for new articles. No spam. Unsubscribe anytime.

By subscribing you agree to receive the newsletter. No spam, and you can unsubscribe anytime.